Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Train identity network from identity with 10% maximum random perturbations #86

Merged
merged 7 commits into from
Sep 20, 2023
Merged

Train identity network from identity with 10% maximum random perturbations #86

merged 7 commits into from
Sep 20, 2023

Commits on Sep 19, 2023

  1. chore(app): whitespace edit

    @rouson
    rouson committed Sep 19, 2023
    Configuration menu
    Copy the full SHA
    eb4d38e View commit details
    Browse the repository at this point in the history

  2. feat(inference_engine_m): add relu activation func

    @rouson
    rouson committed Sep 19, 2023
    Configuration menu
    Copy the full SHA
    a466101 View commit details
    Browse the repository at this point in the history

  3. feat(example): use relu to create linear mapping

    @rouson
    rouson committed Sep 19, 2023
    Configuration menu
    Copy the full SHA
    e1e378d View commit details
    Browse the repository at this point in the history

  4. feat(relu): make differentiable

    @rouson
    rouson committed Sep 19, 2023
    Configuration menu
    Copy the full SHA
    c7fcb68 View commit details
    Browse the repository at this point in the history

  5. feat(trainable_engine):num_outputs type-bound proc

    @rouson
    rouson committed Sep 19, 2023
    Configuration menu
    Copy the full SHA
    451d3db View commit details
    Browse the repository at this point in the history

  6. test(train/relu): preserves identity mapping 

    This commit adds a unit test that verifies that the 'train'
procedure preserves an initially exact network that uses a relu
activation function to represent an mapping of inputs to identical
outputs.
    @rouson
    rouson committed Sep 19, 2023
    Configuration menu
    Copy the full SHA
    e372991 View commit details
    Browse the repository at this point in the history

Commits on Sep 20, 2023

  1. test(train): Adam optimizer trains perturbed identity 

    This commit adds a test in which the use of the Adam optimizer
greatly speeds convergences of a randomly perturbed identity
network with 10% magnitude perturbations. The resulting trained
network maps inputs to nearly identical outputs.
    @rouson
    rouson committed Sep 20, 2023
    Configuration menu
    Copy the full SHA
    6283eee View commit details
    Browse the repository at this point in the history